Discussion of “Pile Aging in Cohesive Soils” by Paul Doherty and Kenneth Gavin

Most published case histories on the growth of pile capacity with time are limited to a rather short period of observations, usually no more than a few weeks. This makes the authors’ 10-year period of observation most welcome. The discusser would like to add the following three case histories: • The Sandpoint case (Fellenius et al. 2004) consists of a 400-mmdiameter, closed-toe, concrete-filled, pipe pile installed into soft clay to a depth of 44m in Idaho. The pile capacity was determined in a dynamic restrike test (CAPWAP) 1 h after the end of driving and in a static loading test 48 days later when piezometermeasurements showed that the pore pressures induced by the driving had dissipated. The pile capacity was again determined in a restrike dynamic test 2,728 days (almost 8 years) later (4 years after the case was published). • The Paddle River case (Fellenius 2008) consists of a series tests in Alberta, Canada, on two 324-mm-diameter, closed-toe, concretefilled, pipe piles driven in clay to depths of 16 and 20 m. One restrike dynamic test was performed less than 1 day after the end of driving, and static loading tests were performed 15, 30, and 1,495 days (4 years) after driving. Most, but not all, of the induced excess pore pressures had dissipated at the 30-day test. • TheKonrad andRoy (1987) case includes capacity determined in static loading tests performed in Quebec, Canada, on a 220-mmdiameter, 7.6-m-long pipe pile driven in soft clay. The tests were carried out 4, 8, 10, and 33 days after the end of driving and on an identical companion pile 4 years after the end of driving. All induced pore pressures had dissipated at the time of the 33-day test. The capacities determined in the three aforementioned cases’ histories are plotted versus time in a logarithmic scale in Fig. 1. The discusser has included the results of the authors’ tests and the authors’ trend line. The Konrad-Roy and the authors’ pile tests (250-mm, 6-m-long concrete piles) can be considered half-scale tests with regard to diameter and/or pile length. To fit the data into the plot, the discusser has scaled up their capacities by a factor of 10. Fig. 1 not only shows an approximately linear logarithmic-scale trend of capacity growth, it also shows all slopes to be somewhat parallel. However, this is amisleading impression. Had the discusser scaled up the results of the half-scale cases by a factor of 20 or 25 instead, their slopeswould have been very different from those of the full-scale pile cases. The use of a logarithmic scale is visually deceiving because the small increase of capacity with time beyond approximately 100 days is exaggerated. This is made clear in Fig. 2, which shows the same data plotted in a linear time scale for the first 200 days. The discusser has added approximate trend lines as dashed lines. The linear-scale plot shows that the process of capacity increase is the result of two processes: pore pressure dissipation and aging, as was also mentioned by the authors. For the case data, the measured time for the pore pressure dissipation ranged from approximately 24 to 50 days after pile construction, which is also when the trend of growth was curved. The small continued growth over the next 100 days is essentially a straight, almost flat line. Of course, normalizing the capacities is a better way to compare the case records than scaling up. The discusser agrees with the authors that such normalization should be to a capacity after the induced excess pore pressures have dissipated, and that the capacity determined for 100 days after the end of construction is a practical choice for the 100% value. Fig. 3 shows this normalization of the records including the linear trend line in Fig. 2 starting at the end of the construction. Beyond the 100-day capacity, this trend line does not agree with the measured capacities and a second linear trend line is necessary to show the trend of the process after the full dissipation of the induced pore pressures. There is a tempting analogy with the

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.